In regard to fab type and excess capacity, I think Intel makes up to the Pentium
120 on 0.6. The 120, 133, 150, 166, and 200 are on 0.35. My understanding is
that early Pentum Pro's were on 0.6, while higher speed 200 MHz parts
moved to 0.35. It is also my understanding that much 0.6 was being converted
to 0.35, and the conversion process was going more smoothly than expected,
thus the quick reductions in high speed Pentium chips.
Moreover, the 486 started on 1.0 micron and moved to 0.8 only with the
486DX-50. The Pentium started on 0.8 micron for the 60/66 and moved to
0.6 for the 90/100 and later the 75 and 120 parts.
The surplus of 0.8 and 0.6 capacity was once thought to presage a
movement back into DRAM for Intel, but then DRAM collapsed thus ending
the utility of that strategy. I would be curious to know what percentage of
Intel fabrication capacity is in use.
Otherwise, keep up the interesting posts. I totally agree IBM is behind the 8-ball in fab capacity/technology. Otherwise, Cyrix would not have had to wait so long for more, and the PowerPC would be that much further ahead.
As rich as IBM is, it hasn't been able to keep pace with Intel manufacturing deployment. Maybe now that it's given up on OS/2 and some other dumb projects it can get its priorities aligned more optimally.
Take care,
Matt |
